TL;DR
Every price and seat limit on this page was verified against claude.com/pricing and Anthropic's support articles on August 21, 2026.
Claude Team costs $20-25 per standard seat per month (1.25x Pro usage) and $100-125 per premium seat (6.25x Pro usage, plus Claude Cowork and Fable 5 inside plan limits). Every seat includes Claude Code. The minimum is now 2 seats, down from 5, and the plan caps at 150 seats. Enterprise is a single seat type at $20 per user per month billed annually, with all usage billed separately at standard API rates; self-serve starts at 20 seats, sales-assisted at 50. For budgeting, Anthropic's own docs report average Claude Code spend of $13 per developer per active day and $150-250 per developer per month across enterprise deployments. A Team premium seat at $100/month annual undercuts an individual Max 5x subscription at the same price while adding 25% more usage and admin controls.
- Solo developer: Claude Max ($100-200/mo) for 5-20x Pro usage, no team overhead
- 2-150 developers: Team plan with mixed Standard ($20/seat annual) and Premium ($100/seat annual) seats
- 20+ developers or compliance requirements: Enterprise at $20/seat + API-rate usage, with SSO, SCIM, audit logs
Anthropic shipped Claude Code into Team and Enterprise plans in August 2025. Before that, every developer on a team needed an individual Max subscription. The CFO complaint was predictable: 14 separate Anthropic charges, no visibility into who used what, no spend controls. The Team plan fixes the billing problem. The Enterprise plan fixes the compliance problem.
Every Claude Plan Compared
| Feature | Pro | Max | Team | Enterprise |
|---|---|---|---|---|
| Monthly price | $20/mo | $100 or $200/mo | $20-125/seat/mo | $20/seat + usage at API rates |
| Claude Code access | Yes | Yes | Yes (all seats) | Yes (all seats) |
| Usage vs Pro | 1x | 5x or 20x | 1.25x (Std) / 6.25x (Prem) | Billed at API rates, no per-seat limits |
| Admin controls | No | No | Yes | Yes |
| SSO | No | No | Yes | Yes |
| SCIM provisioning | No | No | No | Yes |
| Audit logs | No | No | No | Yes |
| Custom data retention | No | No | No | Yes |
| Compliance API | No | No | No | Yes |
| Spending caps | No | No | Org + per-user | Org + per-user |
| Workplace connectors | No | No | Google, Microsoft, Slack, GitHub | Google, Microsoft, Slack, GitHub |
| Minimum seats | 1 | 1 | 2 | 20 self-serve, 50 sales-assisted |
| Maximum seats | 1 | 1 | 150 | Unlimited |
The pricing jumps are nonlinear. A solo developer on Max 5x pays $100/mo for 5x Pro usage. A Team Premium seat costs $100/mo (annual) for 6.25x Pro usage plus admin controls. Since the Team minimum dropped to 2 seats, even a two-person shop gets more usage per dollar from Team Premium than from individual Max subscriptions, plus centralized billing and spend visibility.
Team Plan: Standard vs Premium Seats
The Team plan serves organizations of 2 to 150 people (Anthropic lowered the minimum from 5 seats to 2 during 2026). You can mix Standard and Premium seats based on each member's workload.
| Standard Seat | Premium Seat | |
|---|---|---|
| Annual billing | $20/seat/month | $100/seat/month |
| Monthly billing | $25/seat/month | $125/seat/month |
| Usage vs Pro | 1.25x | 6.25x |
| Claude Code | Included | Included |
| Claude Fable 5 | Usage credits only (not in seat limits) | Up to 50% of weekly limits included |
| Other Claude models | Yes | Yes |
| Extra usage | Available at API rates | Available at API rates |
| Best for | PMs, designers, light users | Developers, power users |
Each team member has their own usage allocation. If one person hits their limit, it does not affect anyone else on the team. Per Anthropic's Claude Code cost docs, the seat allowance resets on a rolling five-hour window plus a weekly window, and it is shared across Claude chat, Claude Code, and Cowork. Through August 31, 2026, Claude Code weekly limits are running 50% above the published standard for Team and other paid plans.
What Both Seats Include
- Claude Code access across terminal, web, desktop, and mobile
- Current Claude models (Opus 5, Sonnet 5, Haiku 4.5; Fable 5 is included in plan limits on premium seats only)
- Workspace connectors: Google Drive, Gmail, Google Calendar, GitHub, Microsoft 365, Slack
- SSO with domain capture and JIT provisioning
- Role-based access controls
- Centralized billing and admin dashboard
The Premium Seat Math
A developer using Claude Code for 4+ hours daily will hit Standard seat limits. Premium seats provide 5x the usage of Standard seats (6.25x vs 1.25x Pro). For a 10-person engineering team where 3 developers are heavy Claude Code users and 7 use Claude for chat and light tasks:
- 7 Standard seats: 7 x $20 = $140/mo (annual)
- 3 Premium seats: 3 x $100 = $300/mo (annual)
- Total: $440/mo for 10 people
Compare that to 10 individual Max 5x subscriptions at $100/mo each ($1,000/mo total), and Team saves $560/mo while adding admin controls.
Enterprise Plan: Usage-Based or Seat-Based
Anthropic restructured Enterprise during 2026. The current plan has one seat type, priced per user per month and billed annually, with usage excluded from the seat fee and billed separately at standard API rates. The claude.com pricing page lists the seat at $20 per user per month. Older Enterprise contracts still carry the legacy structure (Chat seats and Chat + Claude Code seats, or Standard + Premium seats) and transition to the single seat model at contract renewal.
| Current Enterprise (usage-based) | Legacy seat-based Enterprise | |
|---|---|---|
| Seat type | Single all-inclusive seat, $20/user/mo billed annually | Standard + Premium seats (until contract renewal) |
| Usage model | Billed separately at standard API rates | Per-seat usage limits (like Team) |
| Per-seat limits | None (consumption-based) | Yes, per seat allocation |
| Minimum seats | 20 self-serve, 50 sales-assisted | Per original contract |
| Billing | Annual seat fee + usage in arrears | Annual, seat-based |
| Claude Code | Included with every seat, plus Cowork | Chat + Claude Code and Premium seats only |
Enterprise-Only Features
- SSO + SCIM: SAML 2.0 and OIDC-based SSO, with SCIM for automated user provisioning and deprovisioning
- Audit logs: User actions, system events, and data access recorded for compliance
- Custom data retention: Set retention windows to match your organization's data governance policies
- Compliance API: Programmatic access to usage data, activity logs, chat histories, and file content for continuous monitoring
- Analytics API: Aggregated usage and engagement metrics across your organization
- HIPAA readiness: Available through sales-assisted plans with a Business Associate Agreement (BAA)
- Customer-managed encryption keys and custom data retention controls (Anthropic no longer publishes separate per-plan context window figures; current Claude models support up to 1M tokens of context)
On the current Enterprise model there are no individual per-seat usage allocations: consumption bills at API rates as it happens. A developer running a heavy agent team session and someone asking Claude a quick question both pay for exactly what they use. This eliminates wasted capacity from seats that go unused on a given day, at the price of an uncapped bill unless admins set spend limits.
When to Contact Sales
Self-serve Enterprise (minimum 20 seats, credit card or ACH, USD only) covers most organizations. Sales-assisted contracts start at 50 seats. Contact sales if you need invoicing, multi-currency, HIPAA readiness with a BAA, tailored contract terms (MSA, PO, usage commitments), trials, or dedicated customer success management.
Max vs Team Premium: The Individual vs Organization Decision
This is the question engineering managers ask most. Both Max and Team Premium cost roughly $100/mo and both include Claude Code. The difference is organizational infrastructure.
| Max 5x ($100/mo) | Max 20x ($200/mo) | Team Premium ($100/mo annual) | |
|---|---|---|---|
| Usage vs Pro | 5x | 20x | 6.25x |
| Centralized billing | No | No | Yes |
| Spend controls | No | No | Org + per-user caps |
| Usage analytics | Personal only | Personal only | Org-wide dashboards |
| SSO | No | No | Yes |
| Managed policies | No | No | Tool permissions, file access, MCP config |
| Minimum commitment | 1 person | 1 person | 2 seats |
| Can mix seat types | No | No | Yes (Standard + Premium) |
Team Premium at $100/mo (annual) provides 6.25x Pro usage vs Max 5x's 5x. That is 25% more usage for the same price, plus admin controls, plus Fable 5 inside plan limits. With the minimum down to 2 seats, even a two-person team has no reason to stay on individual Max subscriptions.
The exception: if you need 20x Pro usage as an individual and cannot wait for limits to reset, Max 20x at $200/mo is the only option. No Team or Enterprise seat matches that usage tier.
Admin Controls for Claude Code
This is the feature set that blocked enterprise adoption before the Team plan launched. IT departments need visibility into what Claude Code is doing across the organization.
| Control | Team | Enterprise |
|---|---|---|
| Seat management | Self-serve: purchase, allocate, provision | Self-serve + SCIM automated provisioning |
| Spending caps | Org-level and per-user limits | Org-level and per-user limits |
| Usage analytics | Lines accepted, suggestion rate | Lines accepted, suggestion rate, full dashboards |
| Managed policies | Tool permissions, file access, MCP servers | Tool permissions, file access, MCP servers |
| Audit logs | No | User actions, system events, data access |
| Compliance API | No | Programmatic usage data + content access |
| Data retention | Default | Custom retention windows |
Managed Policy Settings
Admins can deploy and enforce settings across all Claude Code users in the organization. This covers three categories:
- Tool permissions: Control which tools Claude Code can execute (shell commands, file operations, browser)
- File access restrictions: Limit which directories and file types Claude Code can read or modify
- MCP server configurations: Approve or restrict which MCP servers users can connect to their Claude Code instances
These policies propagate automatically. When an admin updates a policy, every Claude Code instance in the organization picks it up on next session start. No manual configuration per developer.
Usage Limits and Extra Usage
Claude Code usage on Team and Enterprise plans works differently from individual subscriptions. The billing model depends on your plan type.
| Team (Seat-Based) | Enterprise (Usage-Based) | |
|---|---|---|
| Usage allocation | Per-seat: Std 1.25x Pro, Prem 6.25x Pro | Shared organizational token pool |
| Reset cycle | Rolling 5-hour window + weekly window, per member | Continuous (usage-based) |
| Extra usage | Available at API rates after limit | Billed at API rates (no limit) |
| Unused capacity | Lost at reset (does not roll over) | Pool shared, nothing wasted |
| Avg developer cost | Seat price; $150-250/mo if API-billed | $150-250/mo average at API rates |
Anthropic's Claude Code cost docs (verified August 21, 2026) report that across enterprise deployments, average spend is around $13 per developer per active day and $150-250 per developer per month, with 90% of users staying below $30 per active day. Variance is driven by model choice, how many instances a developer runs, and automation. Agent team sessions use roughly 7x the tokens of a standard session.
Extra Usage on Team Plans
When a Team member hits their included usage limit, they can continue using Claude Code with extra usage billed at standard API rates. Admins control this through spending caps set at the organization level and per-user level. This prevents runaway costs from agent team sessions while keeping developers unblocked.
When to Upgrade: Decision Framework
| Your Situation | Recommended Plan | Why |
|---|---|---|
| Solo developer, heavy usage | Max 5x ($100/mo) | No admin overhead, 5x Pro usage |
| Solo developer, extreme usage | Max 20x ($200/mo) | 20x Pro usage, highest individual tier |
| 2-10 developers, mixed usage | Team (mixed seats) | 3 Premium + 7 Standard = $440/mo annual |
| 10-50 developers, all heavy users | Team (all Premium) | $100/seat annual, centralized billing |
| 50-150 developers | Team or Enterprise | Evaluate if you need SCIM, audit logs, compliance |
| 150+ developers | Enterprise | Team plan caps at 150 seats |
| Regulated industry (HIPAA, SOC2) | Enterprise (sales-assisted) | BAA, custom retention, Compliance API |
| Variable usage across team | Enterprise (usage-based) | Shared pool, no wasted per-seat capacity |
The Migration Path
Individual Pro or Max accounts can migrate to Team or Enterprise organizations. Anthropic handles the transition, and existing usage history carries over. The most common path: start with individual Max subscriptions, move to Team as soon as there are 2 of you, then evaluate Enterprise when compliance requirements appear, team size passes 20, or the 150-seat Team cap approaches.
A 20-person engineering team with 8 heavy Claude Code users and 12 light users. Before Team plan: 20 x Max 5x = $2,000/mo. After Team plan: 8 Premium ($800) + 12 Standard ($240) = $1,040/mo annual. Savings: $960/mo ($11,520/year), plus admin controls and spend visibility.
Frequently Asked Questions
How much does Claude Code cost on the Team plan?
Claude Code is included with every Team plan seat. Standard seats cost $20/mo (annual) or $25/mo (monthly). Premium seats cost $100/mo (annual) or $125/mo (monthly). Premium seats include 6.25x Pro usage vs Standard's 1.25x. Minimum 2 seats required.
What is Claude team pricing per seat?
Claude team pricing runs $20 per member per month billed annually ($25 monthly) for standard seats and $100 billed annually ($125 monthly) for premium seats. Both include Claude Code; premium adds 6.25x Pro usage, Claude Cowork, and Fable 5 inside plan limits. Minimum 2 seats, maximum 150. Verified August 21, 2026.
What is Claude enterprise pricing?
Claude enterprise pricing is a single seat type at $20 per user per month billed annually, plus all usage billed separately at standard API rates. Self-serve requires 20 seats minimum; sales-assisted contracts start at 50 seats and add invoicing and multi-currency. Every seat includes Claude on web, desktop, and mobile, plus Claude Code and Cowork.
What is the difference between Standard and Premium seats?
Both include Claude Code, all models, and workspace connectors. Premium seats provide 5x more usage than Standard seats (6.25x vs 1.25x compared to Pro). Premium is designed for developers running agent teams, multi-file refactoring, and other high-usage workflows.
Should I choose Claude Max or a Team Premium seat?
For the same $100/mo, Team Premium gives you 6.25x Pro usage (vs Max 5x's 5x), plus SSO, spend caps, usage analytics, and managed policies. With the 2-seat minimum, any team of two or more gets the better deal from Team. Max is for solo developers.
Does Enterprise include Claude Code?
Yes. The current Enterprise plan has one seat type at $20 per user per month (billed annually) covering web, desktop, mobile, Claude Code, and Cowork. Usage is not included in the seat fee; it bills separately at standard API rates with no per-seat usage limits.
What context window do Team and Enterprise plans get?
Anthropic no longer publishes separate per-plan context window figures. Current Claude models (Fable 5, Opus 5, Sonnet 5) support up to 1M tokens of context; Haiku 4.5 supports 200K. Older claims of a 200K Team cap and a 500K Sonnet Enterprise tier describe the pre-2026 lineup and are out of date.
Can I mix Standard and Premium seats?
Yes. Admins can assign seat types per user. A common setup: Premium seats for developers using Claude Code daily, Standard seats for PMs, designers, and team members who primarily use Claude chat. Seat upgrades are prorated based on your billing cycle.
What happens when a team member hits their usage limit?
The member can purchase extra usage at standard API rates if the admin has enabled it. Other team members are unaffected. Admins can set per-user spending caps to control extra usage costs.
Your Claude Code Agents Need Better Search
Claude Code's agent teams are only as good as the context they find. WarpGrep pushed Opus 4.6 from 55.4% to 57.5% on SWE-bench Pro by running 8 parallel search calls per turn. Works as an MCP server inside Claude Code on any plan.
Sources
All prices and seat limits verified against these pages on August 21, 2026:
- Claude Help Center: What Is the Max Plan?
- Claude Help Center: Claude Fable 5 on Your Plan
- Claude Help Center: Usage Credits for Paid Claude Plans
- Anthropic: Claude Code and New Admin Controls for Business Plans
- Claude Help Center: What Is the Team Plan?
- Claude Help Center: What Is the Enterprise Plan?
- Claude Help Center: Use Claude Code with Team or Enterprise
- Claude Help Center: Purchasing and Managing Seats
- Claude Pricing Page
- Claude Code Docs: Managing Costs